Two drawings, one block, two sizes
These are 1-2-3 setup blocks: pairs of hardened, ground steel blocks a machinist stands under a workpiece to establish a known height, or bolts to a fixture plate through their hole pattern. The gallery holds two versions of the same technical drawing, with the same line work, the same hole pattern and the same viewpoint. Only the numbers differ.
One is annotated 25 x 50 x 75 MM. The other is annotated L1-2-3 inches, with callouts of 1, 2 and 3 inches on the same three edges.
A third image settles that this is not a printing slip. It shows the card box, printed 1-2-3 BLOCKS, with the text 1-2-3 inch and directly beneath it 25-50-75mm, side by side on the same plain background.
So the listing states both sizes, in three places, and never reconciles them. This is the one question that has to be answered before the blocks are any use, because the two answers are not the same product.
1.6 percent is a conversion, not a machining error
The numbers look close enough to be a rounding habit. They are not rounding. Divide each imperial dimension by the metric one and the same figure comes back three times.
| Axis | 1-2-3 inch | As printed here | Short by | Ratio |
|---|---|---|---|---|
| Thickness | 1.000 in = 25.4 mm | 25 mm | 0.4 mm | 1.0160 |
| Width | 2.000 in = 50.8 mm | 50 mm | 0.8 mm | 1.0160 |
| Length | 3.000 in = 76.2 mm | 75 mm | 1.2 mm | 1.0160 |
All three ratios are 1.0160, to four places. That is not three separate errors; it is one scale factor applied to the whole block. The metric version is a copy at 98.43 percent of true size, and the linear shortfall is 1.575 percent on every axis.
That is exactly what happens when someone converts 1, 2 and 3 inches into round metric numbers and then machines to the round numbers. It is a consistent, deliberate choice, and it is the reason the two drawings can coexist in one gallery: nobody involved treated them as different products.
Volume falls a little faster than length, because three axes shrink together: 93.75 cubic centimetres against 98.32, which is 4.65 percent less steel. At 7.85 grams per cubic centimetre that is about 736 grams against 772, so a pair is roughly 72 grams light. Not a problem in itself, but it is a quick way to tell on arrival which version you were sent, if you have a scale and know what to expect.
What 1.6 percent does to a stack
A single block being 1.2 mm short on its longest axis matters little if you use one block. The error is not the point. The point is that the blocks are designed to be stacked, and the shortfall adds.
| Blocks stacked | Metric total | True 1-2-3 total | Short by | In inches |
|---|---|---|---|---|
| 2 | 150.0 mm | 152.4 mm | 2.4 mm | 0.094 in |
| 3 | 225.0 mm | 228.6 mm | 3.6 mm | 0.142 in |
| 4 | 300.0 mm | 304.8 mm | 4.8 mm | 0.189 in |
| 5 | 375.0 mm | 381.0 mm | 6.0 mm | 0.236 in |
| 6 | 450.0 mm | 457.2 mm | 7.2 mm | 0.283 in |
Six blocks are 7.2 mm short of the height their name promises. If you are building a stack to reach a known height and you calculate that height in inches, you are 0.283 inch out before you have cut anything.
In the other direction: if you are working to round metric numbers and you want 225 mm, these blocks give it in three. That is a perfectly good way to work, and it is what the metric version is for. The danger is buying them for the name and using them for the inch.
The heights are no longer round numbers
A 1-2-3 block earns its place in a toolroom because it offers three known heights that a machinist can hold in their head. Stand it on one face and it is an inch; on another, two; on another, three. The metric version removes that property.
| Nominal | In inches | Nearest round value | Error |
|---|---|---|---|
| 25 mm | 0.984 in | 1 in | 0.016 in |
| 50 mm | 1.969 in | 2 in | 0.031 in |
| 75 mm | 2.953 in | 3 in | 0.047 in |
None of the three is a round inch and none is a round half inch. You cannot touch off against one and add a round number. You have to carry 0.984, 1.969 and 2.953 in your head, or in a notebook, or in the offsets. That is a small tax, but it is paid every time the block is used, and it is exactly the convenience the name was selling.
The hole pattern is the one thing that checks out
It is worth saying what is right, because the rest of the page is thin. Counting the openings on either dimension drawing gives fifteen on the largest face, five along the top edge and three on the end. Fifteen plus five plus three is twenty-three, which is exactly what the title promises and exactly what the convention calls for.
The convention for a 1-2-3 block is twenty-three holes per block, of which five are tapped and eighteen are plain. The photographs bear that out: several holes show cut threads with a visible start chamfer, and the rest are smooth bores.
So the hole count is honest. What is missing is the callout. The five tapped holes are the reason the block is useful for fixturing, and the thread is printed nowhere on the page, in either drawing or in the text.
What the five tapped holes need and do not get
A 1-2-3 block is bought partly as a hole pattern. The tapped holes are where the clamp screw goes, and the plain holes are where the dowel or the through-bolt goes. Without the thread callout you cannot order the bolts, cannot design the fixture plate, and cannot tell whether the holes are through or blind.
The common callout on this class of block is a fractional inch thread, but the range across suppliers is wide enough that guessing is not reasonable, and the drawing marks none of the twenty-three holes as tapped. Ask for the thread, the depth, and which five positions carry it. A drawing with the tapped holes marked is a two-minute request and it prevents a fixture being designed around the wrong pattern.
Hardness, and why it is not a formality
The convention for a block of this kind is hardened and ground steel, commonly quoted in the range of 56 to 60 HRC. Nothing on the page states a hardness, a material grade, or a test method.
This is the number that decides whether the block survives its first month. A soft block takes a dent from a clamp screw, a dropped cutter or a chip trapped under the workpiece. A dented face is no longer flat, and a block that is not flat sits on three corners. The resulting error is small, local and invisible until a part comes out out-of-parallel and nobody can find out why.
Case hardening and through hardening both appear in this product class and they behave differently under a clamp. A case-hardened block that has been ground through its case in one spot is soft there, and there is no way to see it. Ask for the hardness by scale and window, and ask for the case depth if the answer is a case-hardened grade.
Flatness, parallelism, pair matching
These three are the product. The steel is incidental; what you are buying is geometry, and none of the three appears anywhere on the page.
Flatness matters because a setup block is a reference surface. An error transfers directly into the part and cannot be compensated by touching off differently, the way a dimensional error can.
Parallelism and squareness are why a ground block costs more than a piece of key stock cut to length. Ask not whether the block is square but to what figure, over what length, measured how.
Pair matching matters because these are sold as a pair, so the pair is the unit of accuracy. Two blocks that differ in height by 0.02 mm give a stack that is 0.02 mm wrong every time either one is used, and the error changes depending on which way up they are. Ask for the permitted difference inside the pair, and whether the pairs in a shipment are matched to each other or only internally.
Two blocks and a card box
The title says two pieces, and the box photograph shows a plain printed card box. That is the whole kit.
This is worth noticing because the same product is sold elsewhere with a moulded insert, four socket-head cap screws and a hex key in dedicated pockets. Those parts are what let you bolt the blocks to a fixture plate on the day they arrive. Without them, the first job needs a trip to the bolt bin, and the box gives the blocks no protection in transit or on a shelf.
Neither arrangement is wrong, but they are different prices for a reason, and the page does not say which one this is beyond the cardboard in the photograph. Ask what is in the box.
An inspection you can run in an hour
Everything that matters here can be checked on a surface plate with equipment most shops already have, and it is worth doing on the first pair rather than the tenth order.
Measure the three dimensions with a micrometer that reads to a micron. You are looking for 25.4, 50.8 and 76.2, or for 25, 50 and 75. Either answer is fine; a mixture, or a figure between them, is not.
Weigh one block. Around 772 grams says imperial; around 736 says metric. A kitchen scale is not good enough, but a parcel scale is, and the two answers are 36 grams apart, which is well outside any scale error at that range.
Check flatness by wringing the block onto a surface plate, or with a straightedge and feeler if that is all you have, on all six faces. Check the pair by standing both blocks on the same face and running a dial indicator across the top of one against the other; the difference is the pair matching you actually received.
Try a bolt in each of the tapped holes. If a known thread starts cleanly and runs to depth in five holes, the callout matches. If it binds in the first thread, stop and talk to the supplier before the block goes anywhere near a fixture.
What to put in the enquiry
| What a block needs | What is published here | Why it matters |
|---|---|---|
| Material grade | nothing | Decides whether it dents rather than wears |
| Hardness | nothing | Convention is a hardened, ground block, commonly 56 to 60 HRC |
| Dimensional tolerance | nominal only | Nominal without a band tells you nothing about the block you receive |
| Flatness per face | nothing | A setup block is a reference surface; flatness transfers straight into the part |
| Parallelism and squareness | nothing | This is the entire reason to buy a ground block rather than cut bar stock |
| Pair matching | nothing | Sold as a pair, so the pair is the unit of accuracy |
| Thread size and depth | nothing | Without it you cannot order the bolts or start the fixture design |
| Which of the 23 holes are tapped | nothing | Five are tapped by convention, but which five is not stated |
| Coating or finish | nothing | Bare steel blocks rust on a shelf in a damp workshop |
| Weight, net and gross | nothing | Freight and bench handling |
Most of that list is a single email. The answers either exist on a drawing the supplier already has, or they do not, and which of those it is tells you as much as the answers themselves.
| Question | Why |
|---|---|
| Are the blocks 25.4 x 50.8 x 76.2 mm or 25 x 50 x 75 mm | The gallery states both, at three separate places |
| What is the tolerance on each of the three dimensions | Nominal sizes are not a specification |
| What hardness, on which scale, and tested how | Hardness is what keeps the faces flat under a clamp |
| What is the flatness per face, and measured by what method | Unmeasurable claims are not worth paying for |
| What is the parallelism and squareness figure | That is the product; the steel is incidental |
| What is the permitted height difference inside the pair | Two blocks that differ become one wrong height |
| What thread is in the five tapped holes, and how deep | You cannot order bolts without it |
| Which five holes are the tapped ones | Five of twenty-three, and the drawing does not mark them |
| Is the steel coated, and with what | Bare hardened steel rusts |
| What is in the box besides the blocks | The photographs show a card box, no case and no hardware |
| What is the net weight of the pair | Confirms the material and settles the freight |
When the metric blocks are the right buy
None of the above means the blocks are bad. It means they are a different product from the one the name suggests.
If your work is metric, if 25, 50 and 75 mm are the heights you actually want, and if you are using them as spacers and stops rather than as a traceable height standard, then a block that is 1.6 percent under the imperial figure is irrelevant. It is a solid, hardened, ground block with the right hole pattern at a low price, and the arithmetic above never enters your day.
If your work is imperial, if you stack them, or if you use them to establish a height you will then measure against, then 1.2 mm per block is a real error that compounds, and the name is actively misleading. In that case pay for the imperial blocks, or budget 1.2 mm of shim per block and remember to use it.
Beside the parallel set guide
The companion piece, one listing, three products, deals with a different problem on a neighbouring product: a page whose photographs show 1-2-3 blocks while its parameter block specifies PB153 parallel sets in twenty-four pieces, and the height rule hidden in those columns.
That guide ends by listing six numbers the page never states, among them hardness, flatness, pair matching and hole threads. This product page does not fill any of those gaps. What it does do is answer a question the other one could only raise: which of the two size conventions you are looking at. Here the answer is in the gallery, twice, and it disagrees with itself.
Frequently asked questions
Are these blocks 1-2-3 inch or 25-50-75 mm?
The listing states both. One dimension drawing is annotated 25 x 50 x 75 MM and a second version of the same drawing is annotated 1-2-3 inches with 1, 2 and 3 inch callouts. The box photograph prints both figures side by side. Ask the supplier which one is shipped, because they are not the same block.
How much smaller is the metric version?
Exactly 1.575 percent on every axis, which is a scale factor of 1.0160. In absolute terms that is 0.4 mm on the inch, 0.8 mm on the two inch and 1.2 mm on the three inch. All three ratios are identical, so it is one conversion applied to the whole block rather than three separate errors.
Does the difference matter in practice?
It matters if you stack the blocks or use them to establish a height in inches. Three blocks are 3.6 mm short of nine inches and six blocks are 7.2 mm short. It does not matter if your work is metric and 25, 50 and 75 mm are the heights you want anyway.
Does it really have 23 holes?
The count checks out. The drawings show fifteen openings on the largest face, five along the top edge and three on the end, which totals twenty-three, matching both the title and the convention for a 1-2-3 block. By convention five of them are tapped and eighteen are plain, and the photographs do show a mix of threaded and smooth bores.
What thread are the tapped holes?
Not stated anywhere, in either drawing or in the text. That is the single most useful missing number on the page, because the tapped holes are where the clamp screw goes. Ask for the thread, the depth, and which five of the twenty-three positions are tapped.
What hardness should a block like this have?
The convention for a hardened and ground 1-2-3 block is commonly quoted at 56 to 60 HRC. Nothing on this page states a hardness, a steel grade or a test method. A soft block dents under a clamp and a dented face no longer sits flat, so ask for the figure by scale and window.
What comes in the box?
Two blocks and a plain printed card box, judging from the photographs. No moulded case, no cap screws and no hex key appear. The same product is sold elsewhere with a fitted insert and mounting hardware, so ask what is included before comparing prices.
Key specifications
Frequently asked questions
- Are these blocks 1-2-3 inch or 25-50-75 mm?
- The listing states both. One dimension drawing is annotated 25 x 50 x 75 MM and a second version of the same drawing is annotated 1-2-3 inches with 1, 2 and 3 inch callouts. The box photograph prints both figures side by side. Ask the supplier which one is shipped, because they are not the same block.
- How much smaller is the metric version?
- Exactly 1.575 percent on every axis, which is a scale factor of 1.0160. In absolute terms that is 0.4 mm on the inch, 0.8 mm on the two inch and 1.2 mm on the three inch. All three ratios are identical, so it is one conversion applied to the whole block rather than three separate errors.
- Does the difference matter in practice?
- It matters if you stack the blocks or use them to establish a height in inches. Three blocks are 3.6 mm short of nine inches and six blocks are 7.2 mm short. It does not matter if your work is metric and 25, 50 and 75 mm are the heights you want anyway.
- Does it really have 23 holes?
- The count checks out. The drawings show fifteen openings on the largest face, five along the top edge and three on the end, which totals twenty-three, matching both the title and the convention for a 1-2-3 block. By convention five of them are tapped and eighteen are plain, and the photographs do show a mix of threaded and smooth bores.
- What thread are the tapped holes?
- Not stated anywhere, in either drawing or in the text. That is the single most useful missing number on the page, because the tapped holes are where the clamp screw goes. Ask for the thread, the depth, and which five of the twenty-three positions are tapped.
- What hardness should a block like this have?
- The convention for a hardened and ground 1-2-3 block is commonly quoted at 56 to 60 HRC. Nothing on this page states a hardness, a steel grade or a test method. A soft block dents under a clamp and a dented face no longer sits flat, so ask for the figure by scale and window.
- What comes in the box?
- Two blocks and a plain printed card box, judging from the photographs. No moulded case, no cap screws and no hex key appear. The same product is sold elsewhere with a fitted insert and mounting hardware, so ask what is included before comparing prices.